skills/qwenlm/qwen-code/prepare-pr/Gen Agent Trust Hub

prepare-pr

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from branch diffs and logs, creating a surface for indirect prompt injection.
  • Ingestion points: Reads data from git diff, git log, and e2e-report.md (SKILL.md).
  • Boundary markers: Absent. The skill lacks instructions to distinguish between data and potentially malicious embedded instructions.
  • Capability inventory: The skill uses run_shell_command, read_file, and write_file (SKILL.md).
  • Sanitization: None. Content from external sources is used directly to generate the PR body without escaping or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 08:39 AM
Security Audit — agent-trust-hub — prepare-pr